torchlambda

torchlambda copied to clipboard

When I execute **torchlambda layer ./model.ptc --destination "model.zip"** I get the error below: torchlambda:: Started packing Torchscript model to /home/ubuntu18/torchlambda/model.zip. Traceback (most recent call last): File "/home/ubuntu18/.local/bin/torchlambda", line 8, in...

RoyIronGrey

Hi, Would you be able to release a Docker image for Pytorch 1.9?

Pierogi314

10gb lambda

1

So aws lambda now support up to 10gb and increases computational capacity in relation with the mem allocation. I was running inference with 3gb allocation and compared with 10gb but...

Emveez

Following the resnet18 demo. ### First error encountered: `{"errorMessage":"Required field: \"data\" is not string.","errorType":"InvalidJSON", "stackTrace":[]}` Resolved by adding `data = json.dumps(data)` after flattening img in `create_payload.py`. ### Second error: `...

life-efficient

What to do about large models that won't fit into a layer?

2

Is it possible to use an S3 bucket when the model won't fit directly into a lambda layer? Resnet18 is quite small and not typical of DL models these days.

rbunn80110

Python Version 3.7+?

1

First off, I would like to say thank you so much for building this! It has been incredibly easy to use and I am looking forward to testing this out...

lance-miles

Semantic segmentation prediction in output.json

3

Hi, Is it possible to return base64 encoded image segmentation result in output.json? This is the code I came up with after googling ``` auto output = module.forward({normalized_tensor}).toTensor(); torch::Tensor outputTensor...

sagarr

Top k predictions in output?

1

Hello, thanks for the great library! Is it possible to return the top k predictions in the output for the ResNet18 image classification example? And is it possible to include...

mihow

I am getting the following error when I deploy it to lambda: `{"errorMessage": "Bad handler 'torchlambda': not enough values to unpack (expected 2, got 1)", "errorType": "Runtime.MalformedHandlerName", "requestId": "1bbef602-68e0-4251-8e84-636db1c5d1e8", "stackTrace":...

kyuheejo